TY - GEN
T1 - Expectation–maximization algorithm with a nonlinear kalman smoother for MEG/EEG connectivity estimation
AU - Subramaniyam, Narayan Puthanmadam
AU - Tronarp, Filip
AU - Särkkä, Simo
AU - Parkkonen, Lauri
PY - 2017
Y1 - 2017
N2 - Current techniques to determine functional or effective connectivity from magnetoencephalography (MEG) and electroencephalography (EEG) signals typically involve two sequential steps: 1) estimation of the source current distribution from the sensor data, for example, by minimum-norm estimation or beamforming, and 2) fitting a multivariate autoregressive (MVAR) model to estimate the AR coefficients, which reflect the interaction between the sources. Here, we introduce a combination of the expectation–maximization (EM) algorithm and a nonlinear Kalman smoother to perform joint estimation of both source and connectivity (linear and nonlinear) parameters from MEG/EEG signals. Based on simulations, we show that the proposed approach estimates both the source signals and AR coefficients in linear models significantly better than the traditional two-step approach when the signal-to-noise ratio (SNR) is low (≤1) and gives comparable results at higher SNRs (>1). Additionally, we show that nonlinear interaction parameters can be reliably estimated from MEG/EEG signals at low SNRs using the EM algorithm with sigma-point Kalman smoother.
AB - Current techniques to determine functional or effective connectivity from magnetoencephalography (MEG) and electroencephalography (EEG) signals typically involve two sequential steps: 1) estimation of the source current distribution from the sensor data, for example, by minimum-norm estimation or beamforming, and 2) fitting a multivariate autoregressive (MVAR) model to estimate the AR coefficients, which reflect the interaction between the sources. Here, we introduce a combination of the expectation–maximization (EM) algorithm and a nonlinear Kalman smoother to perform joint estimation of both source and connectivity (linear and nonlinear) parameters from MEG/EEG signals. Based on simulations, we show that the proposed approach estimates both the source signals and AR coefficients in linear models significantly better than the traditional two-step approach when the signal-to-noise ratio (SNR) is low (≤1) and gives comparable results at higher SNRs (>1). Additionally, we show that nonlinear interaction parameters can be reliably estimated from MEG/EEG signals at low SNRs using the EM algorithm with sigma-point Kalman smoother.
KW - EEG
KW - Expectation-maximization
KW - Functional connectivity
KW - MEG
KW - Nonlinear Kalman smoother
UR - http://www.scopus.com/inward/record.url?scp=85021750408&partnerID=8YFLogxK
U2 - 10.1007/978-981-10-5122-7_191
DO - 10.1007/978-981-10-5122-7_191
M3 - Conference article in proceedings
AN - SCOPUS:85021750408
SN - 9789811051210
T3 - IFMBE Proceedings
SP - 763
EP - 766
BT - EMBEC and NBC 2017 - Joint Conference of the European Medical and Biological Engineering Conference EMBEC 2017 and the Nordic-Baltic Conference on Biomedical Engineering and Medical Physics, NBC 2017
PB - Springer
T2 - Joint Conference of the European Medical and Biological Engineering Conference, EMBEC 2017 and Nordic-Baltic Conference on Biomedical Engineering and Medical Physics, NBC 2107
Y2 - 11 June 2017 through 15 June 2017
ER -